Donor Advised Funds are Swell. We Should Still Regulate Them Lots.
This weekend the NY Times ran a long story on donor advised funds (“DAFs,” which most people pronounce like the first syllable of Daffy Duck). Since the reporting relied in great measure on quotes and background from my collaborators Ray Madoff and Roger Colinvaux, I have essentially only good things to say about the article. You should read it. Fellow blogger Daniel Hemel is not so sanguine. He thinks the tone of the article, at least, missed some of the positives of DAFs, and oversold the negatives. So herein, some thoughts on Daniel’s thoughts.
First, recall that the term “DAF” is sometimes confusingly used as shorthand for two different things. Technically, a DAF is an investment account owned and invested by a charitable organization. This managing entity is known as the DAF sponsor. Contributors to the account retain the right to “advise” the sponsor, though sponsors that ignored this “advice” would quickly fold. Many people also refer to entities that do little other than manage individual DAF accounts as “donor advised funds,” leaving off the “sponsor.”
Daniel is right that DAFs do some things quite well. They can create economies of scale in processing illiquid donated assets, and can help taxpayers bunch their donations (if you think avoiding Congress’ new floors on deductible contributions are a good thing). I’m more skeptical of his argument that they help people to decide to give away their money. The DAF might make it easier to decide that you’ll give away money some day, but that just postpones your vexing decision about where to give. We have no evidence that DAF’s get more dollars to charities, or get them there any sooner. In IRS data, about 20% of DAF sponsoring organizations reported zero dollars spent per year.
More fundamentally, the pros Daniel points to are all perfectly achievable without also accepting the negatives the Times highlighted. For me, DAFs raise two fundamental (and related) concerns: timing and regulatory arbitrage. The arbitrage point covers a lot of ground, and generally is about the problem that the existence of a DAF rests on a legal fiction. Tax law treats an organization that sponsors at least 5 donor-advised fund accounts as a “public charity,” which is to say it essentially assumes that the organization, not the donor, controls the money. Yet in fact the whole point of the DAF structure is that the donor retains control of their own account in every meaningful respect (except the ability to convert the money back to non-charitable uses).
When tax law does not match economic reality, you have a gold mine for tax advisors. Consider the example of the new excise tax on university endowments. The tax applies to investment funds held by colleges & universities, as well as by entities they “control.” At the recent ABA Tax Section May Meeting, lawyers urged the Treasury to rule that a DAF holding a university’s account wasn’t an entity “controlled” by a contributing university, since the university does not control the sponsoring organization. Yet the university clearly does control its own individual DAF account. With the simple expedient of moving their unrestricted funds to DAFs, universities could effectively repeal the new excise tax.
The biggest arbitrage area right now, noted with some indifference by Daniel, is disclosure. Uniquely among charities, private foundations — organizations that get their money from just a handful of sources, usually one family — must declare publicly whom they support. DAFs don’t. So any existing private foundation can whitewash its donation list simply by first routing its money through a DAF intermediary, in effect repealing the rules that require transparency for family-controlled philanthropic wealth. Daniel poo-poos this as a problem, arguing that a DAF can’t give its money directly to a PAC or 501(c)(4) with certain kinds of political expenditures. Yet as he well knows, it is trivial for a c(4) organization to use a related c(3) to avoid this problem.
Take an organization I just made up, which we could call the National Stifle Association (“NSA”), a c(4). NSA has a related charity, the NSA Fund, with $50m in annual membership revenues, which it uses to cover rent and payroll. A donor — I’ll call him “X. Torshin” — wants to spend $10m on a political ad buy. His DAF gives $10m to the NSA Fund, being careful not to “earmark” it for political spending — but perhaps his friend Mariia suggests some ways his gift might best be used, hint hint. NSA Fund uses this $10m to pay its rent. In turn, it redirects $10m worth of membership dues from rent to the NSA. Voila! Mr. X has funded an independent expenditure or issue ad, and no one knows it was him.
For what it’s worth, Treasury is currently considering regulations on whether private foundations can make distributions to donor-advised funds (or, more technically, whether distributions to or from a DAF should count the same as money to or from other charities). So conceivably this is a loophole that could be closed by regulation.
Now, assuming you still have the patience for all this, a few words about timing. The Times story emphasizes that donors can take a full charitable contribution deduction immediately for a contribution to a DAF, but the DAF need never spend the money. Ever. And, of course, it’s in the financial interests of the so-called commercial DAF sponsors, such as Fidelity and Goldman Sachs, to keep funds in their accounts as long as possible (community foundations are also big DAF sponsors, but often have boards made of nearby charity leaders, and so typically have stronger incentives to get funds out the door).
Here again, Daniel is relatively unbothered, arguing that it’s unclear why we’d want to force firms to spend money sooner rather than later. He cites a brief but well-known essay from Michael Klausner for this point. I have a lot to say about the analysis in Klausner’s piece — you can read an exhaustive take here.
Briefly, my argument is that we need rules to get charities to spend money faster because other legal rules, as well as a set of economic incentives, tend to push them to spend slower than society should want. State laws and the federal tax system encourage slower spending. Managers have incentives to keep spending on things other than their salaries low (or, for DAFs, spending on things other than management fees). And each individual donor or organization may fail to account for the positive spillover benefits spending sooner creates for other firms. Maybe all these claims are wrong, but pointing vaguely in the direction of Klausner and arching an eyebrow isn’t a very satisfying response.
Very long story short, we can accept everything positive Daniel had to say about DAFs, while still thinking they are broken and badly need at least regulatory patching, if not a legislative reworking. Closing all the regulatory loopholes DAFs allow, and making sure that dollars deducted are actually spent in a timely way, could all go hand in hand with a thriving DAF industry.
Post-script: I have one more quibble with Daniel (and with my colleague Jake Brooks). Daniel asserts that the charitable contribution deduction is the only tax benefit DAFs provide, but that isn’t so. A DAF, like any charity, is not taxed on its investment income. A donor who stows their money in a DAF, rather than another investment account, pays no tax on the dividends and realized gains the account produces. That’s especially important if the investor wants to switch from an underperforming investment to a better one. Usually, that switch would carry a realization of accrued gains, a kind of tax exit barrier. DAFs eliminate this barrier. At the margin, we should expect that the economic costs of the exit barrier will equal the tax it imposes. So, while the investor might pay the same amount of tax investing in her own account and in a DAF, her after-tax returns in the DAF will be higher.
